matplotlib-users Mailing List for matplotlib (Page 11)

The source code for radar_chart.py located in the Matplotlib docs at 

http://matplotlib.sourceforge.net/examples/api/radar_chart.html?highlight=radar%20chart

does not work. Seems to be problem with legend.

Can someone tell me what's wrong with the source code so I can correct my
copy?

-- 
View this message in context: http://old.nabble.com/radar_chart-in-matplotlib-doc-does-not-work-tp29177027p29177027.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Since no-one has replied to this, let me see if I can come up with an idea.

Assuming you are using Linux or a Mac, I wonder if it is somehow possible to
"save" a .ps file to a postscript device?  I have never had to do any Linux
magic with CUPS, so maybe this isn't possible.  Anybody else have any
thoughts?

Ben Root


On Tue, Jul 13, 2010 at 6:27 AM, sa6113 <s.p...@gm...> wrote:

>
> Dear all,
>
> I want to print the plotted curves to a printer NOT to a file.
> if you know somthing such as "QwtPlotPrintFilter " class in Qwt.
>
> Which module help me?
>
> --
> View this message in context:
> http://old.nabble.com/print-to-aprint-device%21-tp29149286p29149286.html
> Sent from the matplotlib - users mailing list archive at Nabble.com.
>
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

On Thu, Jul 15, 2010 at 8:41 AM, Waléria Antunes David
<wal...@gm...> wrote:
> Hi all,
>
> I have a code base so that:
>
>
> from pylab import *
> x = arange (3000,3400)
> y = -108 * (3.0e14 ** 2)/x**2
>
> pylab..title("Teste")
> pylab.savefig("imagem.png")
> plot(x, y)
>
>
> Well.... the values of the function range(3000,3400) are in Hz......i need
> to pass GHz which would be in scientific notation as follows bellow:
>
> 3000 Hz = 3,0 × 10-6 GHz
> 3400 Hz = 3,4 x 10-6 Ghz
>
> How do I make the graph x-axis is shown in figures
> scientific notation, for this currently so
>
> 3000,3050,3100,....,3400
>
> in scientific notation is: (3.0e-6,  3.4e-6)

One way is to just change the values in the GHz and plot them:

plot(x/1e9, y)
# Need to change some limits so that they show up in scientific notation:
gca().xaxis.get_major_formatter().set_powerlimits((-5,5))

The other way is make a custom formatter that changes the values of the ticks:

def fmt_ghz(val, pos=None):
    return '%g' % (val / 1e9)

plot(x, y)
gca().xaxis.set_major_formatter(FuncFormatter(fmt_ghz))

You can get more information here:
http://matplotlib.sourceforge.net/api/ticker_api.html

Ryan

-- 
Ryan May
Graduate Research Assistant
School of Meteorology
University of Oklahoma

But, i don't know how do...

I tried, but don't, most failed....

On Thu, Jul 15, 2010 at 11:25 AM, Benjamin Root <ben...@ou...> wrote:

>
> On Thu, Jul 15, 2010 at 8:41 AM, Waléria Antunes David <
> wal...@gm...> wrote:
>
>> Hi all,
>>
>> I have a code base so that:
>>
>>
>> from pylab import *
>> x = arange (3000,3400)
>> y = -108 * (3.0e14 ** 2)/x**2
>>
>> pylab..title("Teste")
>> pylab.savefig("imagem.png")
>> plot(x, y)
>>
>>
>> Well.... the values of the function range(3000,3400) are in Hz......i need
>> to pass GHz which would be in scientific notation as follows bellow:
>>
>> 3000 Hz = 3,0 × 10-6 GHz
>> 3400 Hz = 3,4 x 10-6 Ghz
>>
>> How do I make the graph x-axis is shown in figures
>> scientific notation, for this currently so
>>
>> 3000,3050,3100,....,3400
>>
>> in scientific notation is: (3.0e-6,  3.4e-6)
>>
>> Can someone help me?
>>
>> See my attachment....
>>
>> I need to show this values in scientific notation......
>>
>> Waléria
>>
>>
>>
> Waleria,
>
> I believe you are looking to use the ticker API.  Particularly the
> FormatStrFormatter that accepts a sprintf()-like format string to express
> your tick values.  For your particular example, --off the top of my head--
> the string format would be "%.1e".
>
> http://matplotlib.sourceforge.net/api/ticker_api.html
>
> I hope this helps.
>
> Ben Root
>

On Thu, Jul 15, 2010 at 8:41 AM, Waléria Antunes David <
wal...@gm...> wrote:

> Hi all,
>
> I have a code base so that:
>
>
> from pylab import *
> x = arange (3000,3400)
> y = -108 * (3.0e14 ** 2)/x**2
>
> pylab..title("Teste")
> pylab.savefig("imagem.png")
> plot(x, y)
>
>
> Well.... the values of the function range(3000,3400) are in Hz......i need
> to pass GHz which would be in scientific notation as follows bellow:
>
> 3000 Hz = 3,0 × 10-6 GHz
> 3400 Hz = 3,4 x 10-6 Ghz
>
> How do I make the graph x-axis is shown in figures
> scientific notation, for this currently so
>
> 3000,3050,3100,....,3400
>
> in scientific notation is: (3.0e-6,  3.4e-6)
>
> Can someone help me?
>
> See my attachment....
>
> I need to show this values in scientific notation......
>
> Waléria
>
>
>
Waleria,

I believe you are looking to use the ticker API.  Particularly the
FormatStrFormatter that accepts a sprintf()-like format string to express
your tick values.  For your particular example, --off the top of my head--
the string format would be "%.1e".

http://matplotlib.sourceforge.net/api/ticker_api.html

I hope this helps.

Ben Root

Hello,

I would like to know if it is possible to display the following feature
using matplotlib. Say that you have a 2D plot(x,y) with a lot of data. You
plot the data using a 2D solid line. Would be possible to add "some"
circles/squared or whatever to the solid line? When I say "some" I mean only
scattered values, otherwise given the quantity of data you would appreciate
the symbols.

Thanks,

Clear
-- 
View this message in context: http://old.nabble.com/Scattered-symbols-data-tp29173029p29173029.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Hi Friedrich,
Thanks so much for your response! 

> Date: Wed, 7 Jul 2010 20:16:24 +0200
> 
> Since you are 10.5, I think your default compiler is gcc-4.0 if I'm
> not very much mistaken.  Can you please verify that by issuing gcc
> --version?
> 
> I had a private conversation with the "owner" of the previous thread
> about OS X, and it seems that there are two ways to build matplotlib.
> I personally had no luck with the prebuilt version either, but I don't
> remember the issue, I think it wasn't compatible with my gcc-4.2
> compiled freetype2 (I'm on 10.6).
> 
> a) Use make.osx.  This is quite easy, but you have no control over
> where the libraries are installed, and if you do not log it via 2>&1 |
> tee log.txt we will be unable to diagnose.  make.osx installs
> libraries in your system, so be aware of this.
> 
> b) Install the libraries on your own.  This seems to be a little more
> complicated, but it isn't really.  I used libpng 1.4 with a few
> modifications, just a few lines in src/_png.cpp.  (You also may
> already have recent enough versions through your /opt/ path.)
> 
> Sorry, I'm writing this again and again to all the users I tried to
> convice of this way, but I didn't find inclination to do it properly
> together with the developers on the website.
It is indeed true that I have gcc 4.0i686-apple-darwin9-gcc-4.0.1 (GCC) 4.0.1 (Apple Inc. build 5490)
> Did you compile Python 2.6 yourself?  If yes, have you made sure to
> use the --enable-framework switch for configure?

I installed it using a pre-compiled binary for OS X Leopard, so I did not set any options. Perhaps I should build from source and add this option?
> 
> > I've tried EasyInstall, svn, and dmg. The dmg expects Apple Python 2.6 so
> > that's out. For the EasyInstall and svn routes I think I must be missing
> > some external libraries? Below are some snippets of warnings/error messages:
> 
> It would be helpful to have information about the initial diagnostic
> output of the setup.py call, but I don't know if easy_install works
> that way.  The installation routine tells there what libraries are
> found and if you have pkg-config it tells also the version found.
> 
> > from EasyInstall:
> > $ easy_install matplotlib
> >
> >             matplotlib: 0.99.3
> > warning: no files found matching 'MANIFEST'
> > warning: no files found matching 'lib/mpl_toolkits'
> > ld warning: in /opt/local/lib/libfreetype.dylib, file is not of required
> > architecture
> > ld warning: in /opt/local/lib/libz.dylib, file is not of required
> > architecture
> > ld warning: in /opt/local/lib/libfreetype.dylib, file is not of required
> > architecture
> > ld warning: in /opt/local/lib/libz.dylib, file is not of required
> > architecture
> > ld warning: in /opt/local/lib/libpng12.dylib, file is not of required
> > architecture
> > ld warning: in /opt/local/lib/libz.dylib, file is not of required
> > architecture
> > ld warning: in /opt/local/lib/libfreetype.dylib, file is not of required
> > architecture
> > ld warning: in /opt/local/lib/libz.dylib, file is not of required
> > architecture
> > ld: in /opt/local/lib/libxml2.2.dylib, file is not of required architecture
> > for architecture ppc
> 
> I think /opt/ installs are macports or similar?  What kind of "package
> system" for Mac do you have installed?

That is true... I do have Macports installed and /opt/ is for Macports. But for Python, Numpy, and Scipy, I used the pre-compiled binaries for OS X; the matplotlib I am trying to install now is from the subversion repository. I am not sure what you mean by the initial diagnostic output (perhaps output below?), but I used make.osx and specified that zlib, png and freetype be installed in my ~/dev directory as shown in README.osx and make.osx. I also added all the paths recursively to my PATH in case it was using this variable to look for these programs, but my output error still looks like
basedirlist is: []============================================================================BUILDING MATPLOTLIB            matplotlib: 1.0.svn                python: 2.6.5 (r265:79359, Mar 24 2010, 01:32:55)  [GCC                        4.0.1 (Apple Inc. build 5493)]              platform: darwin
REQUIRED DEPENDENCIES                 numpy: 1.4.1             freetype2: 9.20.3
OPTIONAL BACKEND DEPENDENCIES                libpng: 1.2.35               Tkinter: Tkinter: 73770, Tk: 8.4, Tcl: 8.4              wxPython: no                        * wxPython not found                  Gtk+: no                        * Building for Gtk+ requires pygtk; you must be able                        * to "import gtk" in your build/install environment       Mac OS X native: yes                    Qt: no                   Qt4: no                 Cairo: no
OPTIONAL DATE/TIMEZONE DEPENDENCIES              datetime: present, version unknown              dateutil: matplotlib will provide                  pytz: matplotlib will provideadding pytz
OPTIONAL USETEX DEPENDENCIES                dvipng: 1.9           ghostscript: 8.64                 latex: 3.141592               pdftops: 3.02
[...] and then:ld warning: in /opt/local/lib/libfreetype.dylib, file is not of required architectureld warning: in /opt/local/lib/libz.dylib, file is not of required architectureld: in /opt/local/lib/libxml2.2.dylib, file is not of required architecture for architecture ppccollect2: ld returned 1 exit statusld warning: duplicate dylib /opt/local/lib/libz.1.dyliblipo: can't open input file: /var/folders/Yh/Yh3On1j+FXW+r-334Wk-vk+++TI/-Tmp-//ccUX0Ard.out (No such file or directory)error: command 'c++' failed with exit status 1
> > collect2: ld returned 1 exit status
> 
> It doesn't tell which arch it's missing.  I'm a bit confused about
> this "missing architecture".  What arch does your system have?
> 
My architecture is an Intel Core 2 Duo 2 GHz (64-bit). 
> Please be prepared that it may take a while to sort out all the issues
> on our way.  But I'm sure it's worth!

Sorry for my delayed response - I am currently on travel... but I hope this is fixable. If nothing else comes to mind, I will try installing Python 2.6 from source with the option you mentioned?
Thanks so much!Stephen

 		 	   		  
_________________________________________________________________
The New Busy is not the too busy. Combine all your e-mail accounts with Hotmail.
http://www.windowslive.com/campaign/thenewbusy?tile=multiaccount&ocid=PID28326::T:WLMTAGL:ON:WL:en-US:WM_HMP:042010_4

Eric - 

Thank you very much. This is exactly what I was looking to do. I should have realized the levels are boundaries, and my color array was not matched to the number of regions. The need for BoundaryNorm makes total sense. Time to stop blindly hacking. Thanks for your example script.


- steve

On Jul 14, 2010, at 6:18 PM, Eric Firing wrote:

> On 07/14/2010 12:52 PM, Steve McFarlin wrote:
>> Hello,
>> 
>> I am trying to create a color map that maps 18 colors across 50 levels. As an example let say I have three colors [r,g,b] and want everything between 1 an 2 to be r, 3 through 10 to be g, and 11 through 50 to be b. From what I can tell it does not seem to be possible. Currently this is what I have, but it does not seem to work as I assumed.
>> 
>> colorList = [[0.,0.,102./255.],[0,42./255.,217./255.],[0,110./255.,217./255.],[0,178./255.,217./255.],
>>             [0,212./255.,212./255.],[0,217./255.,166./255.],[0,217./255.,0],[149./255.,217./255.,0],
>>             [217./255.,217./255.,0],[217./255.,174./255.,0],[217./255.,131./255.,0],[217./255.,87./255.,0],
>>             [217./255.,0,0],[174./255.,0,0],[140./255.,0,0],[135./255.,0,0],
>>             [105./255.,0,0],[65./255.,0,0]]
>> 
>> levels = [1,2,3,4,5,6,7,8,9,10,12,15,20,25,30,35,40,50]
>> cmap = matplotlib.colors.ListedColormap(colorList, name = 'theColorMap', N = len(colorList))
>> ...
>> m.contourf(x,y,z,cmap=cmap, levels=levels, extend='both')
>> 
>> If the levels array is continuous then it works as expected. With the above settings I get unexpected results, which includes 'ghost contour lines'. The data I am rendering is from a GRIB file from NOAA.
> 
> I think there is some confusion of terminology here, and the "ghost contour lines" are the least of your problems.
> 
> The contourf "levels" are giving boundaries of regions, so with 18 levels, you have 17 regions.
> 
> I suspect that what you want is illustrated by the attached extremely simple example.  (You may or may not want to use the spacing kwarg to colorbar.)
> 
> Eric
> 
>> 
>> Is this possible?
>> 
>> 
>> Thanks,
>> Steve
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by Sprint
>> What will you do first with EVO, the first 4G phone?
>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
> <contourf_boundary_norm.py>------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Ben - 

The problem is occurring on os-x. I am a version behind on the libraries on this machine. With the latest libraries (installed a few days ago on debian) this problem does not occur with the agg backend. I think it is time to sit down and learn the library rather then hack at it. Thanks for your help. 


Steve
On Jul 14, 2010, at 6:17 PM, Benjamin Root wrote:

> Steve,
> 
> Which backend are you using?  For TkAgg, this works:
> 
> m.contourf(x, y, z, levels=levels, cmap=cmap, antialiased=False)
> 
> Ben Root
> 
> 
> 
> On Wed, Jul 14, 2010 at 8:00 PM, Steve McFarlin <st...@st...> wrote:
> Ben - 
> 
> I got ahead of myself my cropping the image. The color mapping is correct. It is just different then what I expected. I thought it would partition the height field data based on the levels array and index into the color array. Turning off antialiasing did not solve the 'ghost lines'. I will play around with some of the arguments. It looks to me as if contour is being called with a line width of 1 and a color of white. Setting these line width to 0 did not make a difference.
> 
> 
> Thanks,
> Steve 
> 
> On Jul 14, 2010, at 5:24 PM, Benjamin Root wrote:
> 
>> Steve,
>> 
>> The ghost lines appear to be an artifact of the anti-aliasing.  In my tests, setting antialiased=False eliminates those lines.
>> 
>> The colormap looks fine to me.  If you doubt it, try turning on the colorbar to see if the values are correctly associated with the proper colors.
>> 
>> Ben Root
>> 
>> 
>> On Wed, Jul 14, 2010 at 6:24 PM, Steve McFarlin <st...@st...> wrote:
>> Ben - 
>> 
>> You can see the image at http://www.surfguru.com/smc/TestRender.png . With a continuous level array [1 .. 18] the image looks like http://www.surfguru.com/smc/TestRender2.png
>> 
>> 
>> Thanks,
>> Steve
>> 
>> On Jul 14, 2010, at 4:12 PM, Benjamin Root wrote:
>> 
>>> Steve,
>>> 
>>> Could you  please attach an example image of what you are seeing?
>>> 
>>> Ben Root
>>> 
>>> On Wed, Jul 14, 2010 at 5:52 PM, Steve McFarlin <st...@st...> wrote:
>>> Hello,
>>> 
>>> I am trying to create a color map that maps 18 colors across 50 levels. As an example let say I have three colors [r,g,b] and want everything between 1 an 2 to be r, 3 through 10 to be g, and 11 through 50 to be b. From what I can tell it does not seem to be possible. Currently this is what I have, but it does not seem to work as I assumed.
>>> 
>>> colorList = [[0.,0.,102./255.],[0,42./255.,217./255.],[0,110./255.,217./255.],[0,178./255.,217./255.],
>>>            [0,212./255.,212./255.],[0,217./255.,166./255.],[0,217./255.,0],[149./255.,217./255.,0],
>>>            [217./255.,217./255.,0],[217./255.,174./255.,0],[217./255.,131./255.,0],[217./255.,87./255.,0],
>>>            [217./255.,0,0],[174./255.,0,0],[140./255.,0,0],[135./255.,0,0],
>>>            [105./255.,0,0],[65./255.,0,0]]
>>> 
>>> levels = [1,2,3,4,5,6,7,8,9,10,12,15,20,25,30,35,40,50]
>>> cmap = matplotlib.colors.ListedColormap(colorList, name = 'theColorMap', N = len(colorList))
>>> ...
>>> m.contourf(x,y,z,cmap=cmap, levels=levels, extend='both')
>>> 
>>> If the levels array is continuous then it works as expected. With the above settings I get unexpected results, which includes 'ghost contour lines'. The data I am rendering is from a GRIB file from NOAA.
>>> 
>>> Is this possible?
>>> 
>>> 
>>> Thanks,
>>> Steve
>>> ------------------------------------------------------------------------------
>>> This SF.net email is sponsored by Sprint
>>> What will you do first with EVO, the first 4G phone?
>>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>>> _______________________________________________
>>> Matplotlib-users mailing list
>>> Mat...@li...
>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by Sprint
>> What will you do first with EVO, the first 4G phone?
>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> 
>> 
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by Sprint
>> What will you do first with EVO, the first 4G phone?
>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first_______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
> 

On 07/14/2010 12:52 PM, Steve McFarlin wrote:
> Hello,
>
> I am trying to create a color map that maps 18 colors across 50 levels. As an example let say I have three colors [r,g,b] and want everything between 1 an 2 to be r, 3 through 10 to be g, and 11 through 50 to be b. From what I can tell it does not seem to be possible. Currently this is what I have, but it does not seem to work as I assumed.
>
> colorList = [[0.,0.,102./255.],[0,42./255.,217./255.],[0,110./255.,217./255.],[0,178./255.,217./255.],
>              [0,212./255.,212./255.],[0,217./255.,166./255.],[0,217./255.,0],[149./255.,217./255.,0],
>              [217./255.,217./255.,0],[217./255.,174./255.,0],[217./255.,131./255.,0],[217./255.,87./255.,0],
>              [217./255.,0,0],[174./255.,0,0],[140./255.,0,0],[135./255.,0,0],
>              [105./255.,0,0],[65./255.,0,0]]
>
> levels = [1,2,3,4,5,6,7,8,9,10,12,15,20,25,30,35,40,50]
> cmap = matplotlib.colors.ListedColormap(colorList, name = 'theColorMap', N = len(colorList))
> ...
> m.contourf(x,y,z,cmap=cmap, levels=levels, extend='both')
>
> If the levels array is continuous then it works as expected. With the above settings I get unexpected results, which includes 'ghost contour lines'. The data I am rendering is from a GRIB file from NOAA.

I think there is some confusion of terminology here, and the "ghost 
contour lines" are the least of your problems.

The contourf "levels" are giving boundaries of regions, so with 18 
levels, you have 17 regions.

I suspect that what you want is illustrated by the attached extremely 
simple example.  (You may or may not want to use the spacing kwarg to 
colorbar.)

Eric

>
> Is this possible?
>
>
> Thanks,
> Steve
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Steve,

Which backend are you using?  For TkAgg, this works:

m.contourf(x, y, z, levels=levels, cmap=cmap, antialiased=False)

Ben Root



On Wed, Jul 14, 2010 at 8:00 PM, Steve McFarlin <st...@st...>wrote:

> Ben -
>
> I got ahead of myself my cropping the image. The color mapping is correct.
> It is just different then what I expected. I thought it would partition the
> height field data based on the levels array and index into the color array.
> Turning off antialiasing did not solve the 'ghost lines'. I will play around
> with some of the arguments. It looks to me as if contour is being called
> with a line width of 1 and a color of white. Setting these line width to 0
> did not make a difference.
>
>
> Thanks,
> Steve
>
> On Jul 14, 2010, at 5:24 PM, Benjamin Root wrote:
>
> Steve,
>
> The ghost lines appear to be an artifact of the anti-aliasing.  In my
> tests, setting antialiased=False eliminates those lines.
>
> The colormap looks fine to me.  If you doubt it, try turning on the
> colorbar to see if the values are correctly associated with the proper
> colors.
>
> Ben Root
>
>
> On Wed, Jul 14, 2010 at 6:24 PM, Steve McFarlin <st...@st...>wrote:
>
>> Ben -
>>
>> You can see the image at http://www.surfguru.com/smc/TestRender.png .
>> With a continuous level array [1 .. 18] the image looks like
>> http://www.surfguru.com/smc/TestRender2.png
>>
>>
>> Thanks,
>> Steve
>>
>> On Jul 14, 2010, at 4:12 PM, Benjamin Root wrote:
>>
>> Steve,
>>
>> Could you  please attach an example image of what you are seeing?
>>
>> Ben Root
>>
>> On Wed, Jul 14, 2010 at 5:52 PM, Steve McFarlin <st...@st...>wrote:
>>
>>> Hello,
>>>
>>> I am trying to create a color map that maps 18 colors across 50 levels.
>>> As an example let say I have three colors [r,g,b] and want everything
>>> between 1 an 2 to be r, 3 through 10 to be g, and 11 through 50 to be b.
>>> From what I can tell it does not seem to be possible. Currently this is what
>>> I have, but it does not seem to work as I assumed.
>>>
>>> colorList =
>>> [[0.,0.,102./255.],[0,42./255.,217./255.],[0,110./255.,217./255.],[0,178./255.,217./255.],
>>>
>>>  [0,212./255.,212./255.],[0,217./255.,166./255.],[0,217./255.,0],[149./255.,217./255.,0],
>>>
>>>  [217./255.,217./255.,0],[217./255.,174./255.,0],[217./255.,131./255.,0],[217./255.,87./255.,0],
>>>
>>>  [217./255.,0,0],[174./255.,0,0],[140./255.,0,0],[135./255.,0,0],
>>>            [105./255.,0,0],[65./255.,0,0]]
>>>
>>> levels = [1,2,3,4,5,6,7,8,9,10,12,15,20,25,30,35,40,50]
>>> cmap = matplotlib.colors.ListedColormap(colorList, name = 'theColorMap',
>>> N = len(colorList))
>>> ...
>>> m.contourf(x,y,z,cmap=cmap, levels=levels, extend='both')
>>>
>>> If the levels array is continuous then it works as expected. With the
>>> above settings I get unexpected results, which includes 'ghost contour
>>> lines'. The data I am rendering is from a GRIB file from NOAA.
>>>
>>> Is this possible?
>>>
>>>
>>> Thanks,
>>> Steve
>>>
>>> ------------------------------------------------------------------------------
>>> This SF.net email is sponsored by Sprint
>>> What will you do first with EVO, the first 4G phone?
>>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>>> _______________________________________________
>>> Matplotlib-users mailing list
>>> Mat...@li...
>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>
>>
>>
>>
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by Sprint
>> What will you do first with EVO, the first 4G phone?
>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first --
> http://p.sf.net/sfu/sprint-com-first_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
>

Ben - 

I got ahead of myself my cropping the image. The color mapping is correct. It is just different then what I expected. I thought it would partition the height field data based on the levels array and index into the color array. Turning off antialiasing did not solve the 'ghost lines'. I will play around with some of the arguments. It looks to me as if contour is being called with a line width of 1 and a color of white. Setting these line width to 0 did not make a difference.


Thanks,
Steve 

On Jul 14, 2010, at 5:24 PM, Benjamin Root wrote:

> Steve,
> 
> The ghost lines appear to be an artifact of the anti-aliasing.  In my tests, setting antialiased=False eliminates those lines.
> 
> The colormap looks fine to me.  If you doubt it, try turning on the colorbar to see if the values are correctly associated with the proper colors.
> 
> Ben Root
> 
> 
> On Wed, Jul 14, 2010 at 6:24 PM, Steve McFarlin <st...@st...> wrote:
> Ben - 
> 
> You can see the image at http://www.surfguru.com/smc/TestRender.png . With a continuous level array [1 .. 18] the image looks like http://www.surfguru.com/smc/TestRender2.png
> 
> 
> Thanks,
> Steve
> 
> On Jul 14, 2010, at 4:12 PM, Benjamin Root wrote:
> 
>> Steve,
>> 
>> Could you  please attach an example image of what you are seeing?
>> 
>> Ben Root
>> 
>> On Wed, Jul 14, 2010 at 5:52 PM, Steve McFarlin <st...@st...> wrote:
>> Hello,
>> 
>> I am trying to create a color map that maps 18 colors across 50 levels. As an example let say I have three colors [r,g,b] and want everything between 1 an 2 to be r, 3 through 10 to be g, and 11 through 50 to be b. From what I can tell it does not seem to be possible. Currently this is what I have, but it does not seem to work as I assumed.
>> 
>> colorList = [[0.,0.,102./255.],[0,42./255.,217./255.],[0,110./255.,217./255.],[0,178./255.,217./255.],
>>            [0,212./255.,212./255.],[0,217./255.,166./255.],[0,217./255.,0],[149./255.,217./255.,0],
>>            [217./255.,217./255.,0],[217./255.,174./255.,0],[217./255.,131./255.,0],[217./255.,87./255.,0],
>>            [217./255.,0,0],[174./255.,0,0],[140./255.,0,0],[135./255.,0,0],
>>            [105./255.,0,0],[65./255.,0,0]]
>> 
>> levels = [1,2,3,4,5,6,7,8,9,10,12,15,20,25,30,35,40,50]
>> cmap = matplotlib.colors.ListedColormap(colorList, name = 'theColorMap', N = len(colorList))
>> ...
>> m.contourf(x,y,z,cmap=cmap, levels=levels, extend='both')
>> 
>> If the levels array is continuous then it works as expected. With the above settings I get unexpected results, which includes 'ghost contour lines'. The data I am rendering is from a GRIB file from NOAA.
>> 
>> Is this possible?
>> 
>> 
>> Thanks,
>> Steve
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by Sprint
>> What will you do first with EVO, the first 4G phone?
>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> 
> 
> 
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
> 
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Steve,

The ghost lines appear to be an artifact of the anti-aliasing.  In my tests,
setting antialiased=False eliminates those lines.

The colormap looks fine to me.  If you doubt it, try turning on the colorbar
to see if the values are correctly associated with the proper colors.

Ben Root


On Wed, Jul 14, 2010 at 6:24 PM, Steve McFarlin <st...@st...>wrote:

> Ben -
>
> You can see the image at http://www.surfguru.com/smc/TestRender.png . With
> a continuous level array [1 .. 18] the image looks like
> http://www.surfguru.com/smc/TestRender2.png
>
>
> Thanks,
> Steve
>
> On Jul 14, 2010, at 4:12 PM, Benjamin Root wrote:
>
> Steve,
>
> Could you  please attach an example image of what you are seeing?
>
> Ben Root
>
> On Wed, Jul 14, 2010 at 5:52 PM, Steve McFarlin <st...@st...>wrote:
>
>> Hello,
>>
>> I am trying to create a color map that maps 18 colors across 50 levels. As
>> an example let say I have three colors [r,g,b] and want everything between 1
>> an 2 to be r, 3 through 10 to be g, and 11 through 50 to be b. From what I
>> can tell it does not seem to be possible. Currently this is what I have, but
>> it does not seem to work as I assumed.
>>
>> colorList =
>> [[0.,0.,102./255.],[0,42./255.,217./255.],[0,110./255.,217./255.],[0,178./255.,217./255.],
>>
>>  [0,212./255.,212./255.],[0,217./255.,166./255.],[0,217./255.,0],[149./255.,217./255.,0],
>>
>>  [217./255.,217./255.,0],[217./255.,174./255.,0],[217./255.,131./255.,0],[217./255.,87./255.,0],
>>
>>  [217./255.,0,0],[174./255.,0,0],[140./255.,0,0],[135./255.,0,0],
>>            [105./255.,0,0],[65./255.,0,0]]
>>
>> levels = [1,2,3,4,5,6,7,8,9,10,12,15,20,25,30,35,40,50]
>> cmap = matplotlib.colors.ListedColormap(colorList, name = 'theColorMap', N
>> = len(colorList))
>> ...
>> m.contourf(x,y,z,cmap=cmap, levels=levels, extend='both')
>>
>> If the levels array is continuous then it works as expected. With the
>> above settings I get unexpected results, which includes 'ghost contour
>> lines'. The data I am rendering is from a GRIB file from NOAA.
>>
>> Is this possible?
>>
>>
>> Thanks,
>> Steve
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by Sprint
>> What will you do first with EVO, the first 4G phone?
>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>
>
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

Ben - 

You can see the image at http://www.surfguru.com/smc/TestRender.png . With a continuous level array [1 .. 18] the image looks like http://www.surfguru.com/smc/TestRender2.png


Thanks,
Steve

On Jul 14, 2010, at 4:12 PM, Benjamin Root wrote:

> Steve,
> 
> Could you  please attach an example image of what you are seeing?
> 
> Ben Root
> 
> On Wed, Jul 14, 2010 at 5:52 PM, Steve McFarlin <st...@st...> wrote:
> Hello,
> 
> I am trying to create a color map that maps 18 colors across 50 levels. As an example let say I have three colors [r,g,b] and want everything between 1 an 2 to be r, 3 through 10 to be g, and 11 through 50 to be b. From what I can tell it does not seem to be possible. Currently this is what I have, but it does not seem to work as I assumed.
> 
> colorList = [[0.,0.,102./255.],[0,42./255.,217./255.],[0,110./255.,217./255.],[0,178./255.,217./255.],
>            [0,212./255.,212./255.],[0,217./255.,166./255.],[0,217./255.,0],[149./255.,217./255.,0],
>            [217./255.,217./255.,0],[217./255.,174./255.,0],[217./255.,131./255.,0],[217./255.,87./255.,0],
>            [217./255.,0,0],[174./255.,0,0],[140./255.,0,0],[135./255.,0,0],
>            [105./255.,0,0],[65./255.,0,0]]
> 
> levels = [1,2,3,4,5,6,7,8,9,10,12,15,20,25,30,35,40,50]
> cmap = matplotlib.colors.ListedColormap(colorList, name = 'theColorMap', N = len(colorList))
> ...
> m.contourf(x,y,z,cmap=cmap, levels=levels, extend='both')
> 
> If the levels array is continuous then it works as expected. With the above settings I get unexpected results, which includes 'ghost contour lines'. The data I am rendering is from a GRIB file from NOAA.
> 
> Is this possible?
> 
> 
> Thanks,
> Steve
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 

Steve,

Could you  please attach an example image of what you are seeing?

Ben Root

On Wed, Jul 14, 2010 at 5:52 PM, Steve McFarlin <st...@st...>wrote:

> Hello,
>
> I am trying to create a color map that maps 18 colors across 50 levels. As
> an example let say I have three colors [r,g,b] and want everything between 1
> an 2 to be r, 3 through 10 to be g, and 11 through 50 to be b. From what I
> can tell it does not seem to be possible. Currently this is what I have, but
> it does not seem to work as I assumed.
>
> colorList =
> [[0.,0.,102./255.],[0,42./255.,217./255.],[0,110./255.,217./255.],[0,178./255.,217./255.],
>
>  [0,212./255.,212./255.],[0,217./255.,166./255.],[0,217./255.,0],[149./255.,217./255.,0],
>
>  [217./255.,217./255.,0],[217./255.,174./255.,0],[217./255.,131./255.,0],[217./255.,87./255.,0],
>            [217./255.,0,0],[174./255.,0,0],[140./255.,0,0],[135./255.,0,0],
>            [105./255.,0,0],[65./255.,0,0]]
>
> levels = [1,2,3,4,5,6,7,8,9,10,12,15,20,25,30,35,40,50]
> cmap = matplotlib.colors.ListedColormap(colorList, name = 'theColorMap', N
> = len(colorList))
> ...
> m.contourf(x,y,z,cmap=cmap, levels=levels, extend='both')
>
> If the levels array is continuous then it works as expected. With the above
> settings I get unexpected results, which includes 'ghost contour lines'. The
> data I am rendering is from a GRIB file from NOAA.
>
> Is this possible?
>
>
> Thanks,
> Steve
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Hello,

I am trying to create a color map that maps 18 colors across 50 levels. As an example let say I have three colors [r,g,b] and want everything between 1 an 2 to be r, 3 through 10 to be g, and 11 through 50 to be b. From what I can tell it does not seem to be possible. Currently this is what I have, but it does not seem to work as I assumed.

colorList = [[0.,0.,102./255.],[0,42./255.,217./255.],[0,110./255.,217./255.],[0,178./255.,217./255.],
            [0,212./255.,212./255.],[0,217./255.,166./255.],[0,217./255.,0],[149./255.,217./255.,0],
            [217./255.,217./255.,0],[217./255.,174./255.,0],[217./255.,131./255.,0],[217./255.,87./255.,0],
            [217./255.,0,0],[174./255.,0,0],[140./255.,0,0],[135./255.,0,0],
            [105./255.,0,0],[65./255.,0,0]]

levels = [1,2,3,4,5,6,7,8,9,10,12,15,20,25,30,35,40,50]
cmap = matplotlib.colors.ListedColormap(colorList, name = 'theColorMap', N = len(colorList)) 
...
m.contourf(x,y,z,cmap=cmap, levels=levels, extend='both')

If the levels array is continuous then it works as expected. With the above settings I get unexpected results, which includes 'ghost contour lines'. The data I am rendering is from a GRIB file from NOAA. 

Is this possible?


Thanks,
Steve

On 07/14/2010 11:41 AM, Stephen Evans wrote:
> Hi,
>
> While testing Psyco V2 to see if it would offer any speed improvements I
> tried it with some applications using matplotlib. Exceptions were raised
> that were easily resolved by replacing calls to min() and max() with
> their numpy equivalents numpy.amin() and numpy.amax() in the matplotlib
> code.
>
> Simply demonstrated by inserting at the beginning of, say, matplotlib's
> examples/api/barchart_demo.py :
>
> import psyco
> psyco.full()
>
> which caused:
>
> Traceback (most recent call last):
>     File "barchart_demo.py", line 29, in<module>
>       ax.set_xticks(ind+width)
>     File "c:\python26\lib\site-packages\matplotlib\axes.py", line 2064,
> in set_xticks
>       return self.xaxis.set_ticks(ticks, minor=minor)
>     File "c:\python26\lib\site-packages\matplotlib\axis.py", line 1154,
> in set_ticks
>       self.set_view_interval(min(ticks), max(ticks))
>     File "c:\python26\lib\site-packages\psyco\builtin.py", line 75, in min
>       return _min(*args)
>     File "c:\python26\lib\site-packages\psyco\builtin.py", line 34, in _min
>       if not iterable:
> ValueError: The truth value of an array with more than one element is
> ambiguous. Use a.any() or a.all()
>
>
> software used:
>
> Python 2.6.5 (r265:79096, Mar 19 2010, 21:48:26) [MSC v.1500 32 bit
> (Intel)] on win32
> Type "help", "copyright", "credits" or "license" for more information.
>   >>>  import psyco
>   >>>  psyco.version_info
> (2, 0, 0, 'final', 0)
>   >>>  import numpy
>   >>>  numpy.version.version
> '1.4.1'
>   >>>  import matplotlib
>   >>>  matplotlib.__version__
> '0.99.3'
>
>
> Psyco V2 is available from: http://codespeak.net/svn/psyco/v2/dist/
>
> Should numpy.min()/numpy.amin() be used on array like objects within
> matplotlib, or is min() adequate ? Ditto max().

When short sequences are involved, min() is much faster than amin(). If 
min() is called only a few times per plot in such cases, using the 
slower function would cause a negligible slowdown.  I'm reluctant to 
change mpl to work around a bug in psyco, though.

When you did make the substitution and do the test, was there a big speedup?

Eric

>
>
> Stephen Evans
>
> (Out of interest I came across numpy ticket #1286 while looking for this
> issue.)
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Hi,

While testing Psyco V2 to see if it would offer any speed improvements I 
tried it with some applications using matplotlib. Exceptions were raised 
that were easily resolved by replacing calls to min() and max() with 
their numpy equivalents numpy.amin() and numpy.amax() in the matplotlib 
code.

Simply demonstrated by inserting at the beginning of, say, matplotlib's 
examples/api/barchart_demo.py :

import psyco
psyco.full()

which caused:

Traceback (most recent call last):
   File "barchart_demo.py", line 29, in <module>
     ax.set_xticks(ind+width)
   File "c:\python26\lib\site-packages\matplotlib\axes.py", line 2064, 
in set_xticks
     return self.xaxis.set_ticks(ticks, minor=minor)
   File "c:\python26\lib\site-packages\matplotlib\axis.py", line 1154, 
in set_ticks
     self.set_view_interval(min(ticks), max(ticks))
   File "c:\python26\lib\site-packages\psyco\builtin.py", line 75, in min
     return _min(*args)
   File "c:\python26\lib\site-packages\psyco\builtin.py", line 34, in _min
     if not iterable:
ValueError: The truth value of an array with more than one element is 
ambiguous. Use a.any() or a.all()


software used:

Python 2.6.5 (r265:79096, Mar 19 2010, 21:48:26) [MSC v.1500 32 bit 
(Intel)] on win32
Type "help", "copyright", "credits" or "license" for more information.
 >>> import psyco
 >>> psyco.version_info
(2, 0, 0, 'final', 0)
 >>> import numpy
 >>> numpy.version.version
'1.4.1'
 >>> import matplotlib
 >>> matplotlib.__version__
'0.99.3'


Psyco V2 is available from: http://codespeak.net/svn/psyco/v2/dist/

Should numpy.min()/numpy.amin() be used on array like objects within 
matplotlib, or is min() adequate ? Ditto max().


Stephen Evans

(Out of interest I came across numpy ticket #1286 while looking for this 
issue.)

I'm embedding a plot in a wxPython panel, and I'd like it to blend in as
much as possible, i.e. the area outside the plot itself should be
transparent.  I can call figure.figurePatch.set_alpha(0.0) to do this under
normal circumstances, but in this case I'm also trying to redraw the plot
periodically in response to a background process.  When I do this,
figure.clear() clears the plot area, but not the axes and labels, so I end
up with all of the old axes superimposed.  Is there a way to completely
erase the background before redrawing?

thanks,
Nat

On 07/14/2010 06:49 AM, Daniel Welling wrote:
> Greetings.
>
> I've recently found that when I replace pcolor with pcolorfast, the
> image will not scale correctly when placed on an axis with a logarithmic
> scale.  It will remain linear, thus not matching the axis range
> whatsoever.  The pcolor plot will still fit nicely in the axis object,
> but the ticks and labels clearly do not match the data.  I've tried
> several things, from changing the axis from linear to log before and
> after using pcolorfast, etc, but pcolorfast artist objects do not seem
> to respond to this.  Again, pcolor acts as one would expect.
>
> Complicating this is that I have made two changes at once: upgrading
> from MPL 99 to 1.0 and switching from pcolor to pcolorfast.  As such, I
> do not know if this a new or old issue (or if it is an issue at- perhaps
> this behavior is on purpose.)
>
> If anyone could shed some light on this, that would be great.
>   Pcolorfast is preferable over pcolor because of the complexity of my
> plots and the speed up gained by using pcolorfast.

The problem is that I never quite finished pcolorfast; it probably needs 
a little more refinement in argument handling, and it needs to detect 
nonlinear axes.  The latter never even occurred to me when I wrote it, 
since that case does not occur in the applications for which I use it.

For now, try using pcolormesh instead of pcolorfast. It will handle log 
axes, and although probably not quite as fast as pcolorfast could be if 
it correctly handled log axes, it will be *much* faster than pcolor.

Eric

>
> Thanks!
> -dw
>
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

On 07/14/2010 01:29 AM, Georges Schutz wrote:
> Hi again,
> While looking for a solution for my autoscale issue (see 13/7/2010
> 16:50) I had a look at the matplotlib (1.0.0) source code and found
> something strange in axes.py
>
> In the Axes class the method margins()(Line 1651) the doc string says
> that using margins() with no arguments should return xmargin, ymargin.
>
> If I look at the code for that situation
> --- axes.py line (1686-1687)
>           if not args and not kw:
>               return self._ymargin, self._ymargin
> -----
> the ymargin is returned twice.
>
> I suppose that this is a bug and should be
>
>     return self._xmargin, self._ymargin

Thank you.  Fixed in 8549, 8550.

Eric

>
> Best regards
>
> Georges Schutz
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

On 2010-07-14 18:51:26 +0200, K.-Michael Aye said:

> On 2010-07-14 18:45:35 +0200, John Hunter said:
> 
>> On Wed, Jul 14, 2010 at 11:38 AM, K.-Michael Aye
>> <kmi...@gm...> wrote:
>> 
>>> Out[12]: 1
>>> 
>>> In [13]: gc.collect()
>>> 
>>> Out[13]: 12
>> 
>> 
>> still not seeing a leak in your data -- you need to report_memory
>> after calling gc collect.  Turn off hold, add an image, call collect,
>> report memory, the repeat several times, each time calling collect and
>> report memory, and report the results.
> 
> Was just following your example, you were nowhere calling collect.
> Here is what you requested:
> 
> In [1]: import gc
> 
> In [2]: import matplotlib.cbook as cbook
> 
> In [3]: data = ones((1500,1500,3))
> 
> In [4]: hold(False)
> 
> In [5]: imshow(data)
> 
> Out[5]: <matplotlib.image.AxesImage object at 0x1c43e50>
> 
> In [6]: gc.collect()
> 
> Out[6]: 12
> 
> In [7]: cbook.report_memory()
> 
> Out[7]: 174540
> 
> In [8]: imshow(data)
> 
> Out[8]: <matplotlib.image.AxesImage object at 0x1c59e90>
> 
> In [9]: gc.collect()
> 
> Out[9]: 0
> 
> In [10]: cbook.report_memory()
> 
> Out[10]: 253400
> 
> In [11]: imshow(data)
> 
> Out[11]: <matplotlib.image.AxesImage object at 0x1c603b0>
> 
> In [12]: gc.collect()
> 
> Out[12]: 0
> 
> In [13]: cbook.report_memory()
> 
> Out[13]: 333360
> 
> In [14]: imshow(data)
> 
> Out[14]: <matplotlib.image.AxesImage object at 0x1c60410>
> 
> In [15]: gc.collect()
> 
> Out[15]: 0
> 
> In [16]: cbook.report_memory()
> 
> Out[16]: 413296
> 


Here are the commands as macro form, for easy cut and paste into pylab:

import gc
import matplotlib.cbook as cbook
data = ones((1500,1500,3))
hold(False)
imshow(data)
gc.collect()
cbook.report_memory()
imshow(data)
gc.collect()
cbook.report_memory()
imshow(data)
gc.collect()
cbook.report_memory()
imshow(data)
gc.collect()
cbook.report_memory()


> 
> 
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Sprint
> What will you do first with EVO, the first 4G phone?
> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first

On 2010-07-14 18:45:35 +0200, John Hunter said:

> On Wed, Jul 14, 2010 at 11:38 AM, K.-Michael Aye 
> <kmi...@gm...> wrote:
> 
>> Out[12]: 1
>> 
>> In [13]: gc.collect()
>> 
>> Out[13]: 12
> 
> 
> still not seeing a leak in your data -- you need to report_memory
> after calling gc collect.  Turn off hold, add an image, call collect,
> report memory, the repeat several times, each time calling collect and
> report memory, and report the results.

Was just following your example, you were nowhere calling collect.
Here is what you requested:

In [1]: import gc

In [2]: import matplotlib.cbook as cbook

In [3]: data = ones((1500,1500,3))

In [4]: hold(False)

In [5]: imshow(data)

Out[5]: <matplotlib.image.AxesImage object at 0x1c43e50>

In [6]: gc.collect()

Out[6]: 12

In [7]: cbook.report_memory()

Out[7]: 174540

In [8]: imshow(data)

Out[8]: <matplotlib.image.AxesImage object at 0x1c59e90>

In [9]: gc.collect()

Out[9]: 0

In [10]: cbook.report_memory()

Out[10]: 253400

In [11]: imshow(data)

Out[11]: <matplotlib.image.AxesImage object at 0x1c603b0>

In [12]: gc.collect()

Out[12]: 0

In [13]: cbook.report_memory()

Out[13]: 333360

In [14]: imshow(data)

Out[14]: <matplotlib.image.AxesImage object at 0x1c60410>

In [15]: gc.collect()

Out[15]: 0

In [16]: cbook.report_memory()

Out[16]: 413296

Greetings.

I've recently found that when I replace pcolor with pcolorfast, the image
will not scale correctly when placed on an axis with a logarithmic scale.
 It will remain linear, thus not matching the axis range whatsoever.  The
pcolor plot will still fit nicely in the axis object, but the ticks and
labels clearly do not match the data.  I've tried several things, from
changing the axis from linear to log before and after using pcolorfast, etc,
but pcolorfast artist objects do not seem to respond to this.  Again, pcolor
acts as one would expect.

Complicating this is that I have made two changes at once: upgrading from
MPL 99 to 1.0 and switching from pcolor to pcolorfast.  As such, I do not
know if this a new or old issue (or if it is an issue at- perhaps this
behavior is on purpose.)

If anyone could shed some light on this, that would be great.  Pcolorfast is
preferable over pcolor because of the complexity of my plots and the speed
up gained by using pcolorfast.

Thanks!
-dw

On Wed, Jul 14, 2010 at 11:38 AM, K.-Michael Aye <kmi...@gm...> wrote:

> Out[12]: 1
>
> In [13]: gc.collect()
>
> Out[13]: 12


still not seeing a leak in your data -- you need to report_memory
after calling gc collect.  Turn off hold, add an image, call collect,
report memory, the repeat several times, each time calling collect and
report memory, and report the results.

JDH